This project utilizes a Super-Resolution Generative Adversarial Network (SRGAN) to enhance the resolution and reduce noise in Head CT scan images. The primary objective is to minimize patient exposure to CT radiation by generating high-quality images from low-dose CT scans. This approach aims to achieve high-resolution CT images comparable to those obtained from higher radiation doses, thus improving patient safety without compromising image quality.
In medical imaging, especially in CT scans, there's a critical balance between image quality and radiation dose. High-dose CT scans provide clearer and more detailed images but expose patients to higher levels of radiation, which can have long-term health risks. This project addresses this issue by using SRGAN to enhance low-dose CT scans, producing high-resolution images that retain diagnostic quality while significantly reducing the radiation dose required.
- Super Resolution: The SRGAN model enhances the spatial resolution of low-dose CT images, making fine anatomical details visible.
- Denoising: The model also effectively reduces noise, which is a common issue in low-dose CT scans, ensuring that the enhanced images are clear and useful for diagnostic purposes.
- DigitalSreeni - https://www.youtube.com/@DigitalSreeni
- Aladdin Persson - https://www.youtube.com/@AladdinPersson